Tuesday, March 06, 2012

Pointers, Not Papers

Having just crammed everything we could of a 10 page paper into 5 pages for a workshop paper, I admit that besides (once again) being annoyed by having to go through this process, I'm not clear for the reasons behind it. 

I understand reasons for page limits.  Some authors would not self-edit unless forced to do so, and a little regulation arguably makes life better for everyone -- especially the reviewers, but also other readers, and even the authors who wouldn't otherwise take care to edit their text.  In some (rare, I think, these days) cases, there might actually be sound logistical publishing reasons for a page limit.  Also, one might argue that, as a matter of fairness, authors should be obliged to have a somewhat tight rather than loose upper bound on what they can write. 

But 5 double-spaced pages?  (Or, for those who use LNCS format, 12 pages, which is roughly the same?)  Page 1 is a basic intro, Page 2 gives relevant background, such as prior work, definitions and terminology, Page 3 (for my papers, generally) provides theoretical results (little or no proofs, of course, for any subject with meat -- at best a high-level sketch and discussion), Page 4 and whatever room I can take from Page 5 for experiments (including plots!) and analysis, and what's left of Page 5 for a brief conclusion and the references.  Anyone who writes papers can tell there's going to be plenty of stuff you have to cut to make this work.  Some of what is cut is just the textured, nuanced commentary that would help the reader;  some of it is actually important.  (Leading to my irrational anger when I get reviewer comments of the form:  "Why didn't you talk about XXX?" when my answer is WE DID, but we had to take it out to meet his arbitrary page limit.)  

Papers these days, I guess, are often just advertisements for the work up on the arxiv.*  Is this the product of a reasoned argument on the right page limit?  I'd like to hear the arguments.**

*  I note some conferences have gone the other way, allowing longer papers since it's all electronic and cost of paper is not really a concern.  See EC this year...

** Right now the only one I know of is that workshop papers for things like HOT papers should be 5-6 pages, so when you submit the 10-14 page version to the regular conference there will be enough "new" material to justify another publication.  I don't find this a good argument, personally.    


Michael Albert said...

"Papers these days, I guess, are often just advertisements for the work up on the arxiv."

True, so why not go the other way -- the conference submission limit is (say) 2 pages. In that space you try to convince the referee that it's worth his/her time to look at the (mandatory) arXiv version.

Michael Mitzenmacher said...

Certainly a possible idea. But then, why limit the size of conferences at all? Give everyone who wants it their 2 pages, put them online, and let people figure out what talks they want to go to.

I know some communities that work that way; I personally haven't enjoyed the few conferences of that type I've gone to.

Michael Albert said...

That's fairly much the traditional model in mathematics (which is where I started, and still largely publish). The tendency in CS to evaluate conference proceedings as comparable to, or in some cases superior to, journal articles would mean that it would be a pretty big paradigm shift for that discipline.

D. Eppstein said...

Conferences that don't limit the page counts have been known to cause problems for authors who want to publish a more polished and carefully refereed version of their paper in a journal, only to find out that they can't because the polish and refereeing doesn't count towards the 30% "new material" that many journals require. So by fixing one problem (too-narrow page limits) you've introduced another one — maybe this is a sign that we're looking at the problem at too low a level of detail and that bigger things are broken about academic publishing than page counts.

David Andersen said...

Hot*: If you can't convince me that your idea is cool and hot in 5 pages, it's not hot. These workshops (should be) about the niftiness of the idea. If it takes you another 9 pages of in-depth evaluation to convince me, then just write the damn conference paper, tyvm, because your idea isn't that hot or provocative.

Overall: A well-selected page limit acts like an editor: It ideally forces people to express themselves more concisely. The problem, of course, is that it's a coarse hammer. There's no one optimal page length for all papers. But most of my papers have benefitted from the time we've taken to chop them down to 14 pages - even if we then got to expand them a little bit for the camera-ready. Asking, "does this {element} really add value to the paper?" is a great question that's typically only answered under duress.

Michael Mitzenmacher said...

David Andersen:

1) I call BS on your 5-page statement. Oh, I might be able to convince I have an IDEA in 5 pages. But I won't be able to convince you it's really actually different than previous ideas, and actually potentially viable, in 5 pages a substantial fraction of the time. (As my outline, I think, argues.)

Networking conferences often have 14-page papers. How about 7-page Hot* papers?

Overall: As my post says, I'm OK with some sort of reasonably well-thought out page limit, as it does force editing. I'm not convinced "14" is the right magic number, but in my experience it's generally quite reasonable.

Mikko Särelä said...

One viewpoint to the matter. At least some single-track workshops try to elicit real discussion. And real discussion requires that people actually have the possibility to read the paper's in advance. One can go through thirty 5-page papers in a day or two, but doing the same for ten pagers takes much longer.

Hence, the requirement to condense the paper to its core idea.

No, I'm not saying this is a good reason for all workshops, but anyhow there it is.

Anonymous said...

In my experience, part of the problem with longer page limits is that reviewers penalize you for submitting a paper that doesn't use all the pages. The result is that the longer page limit forces you to pad, and in my opinion (as both an author and a reviewer) it makes the papers worse.

Why reviewers do this I don't know...

David Andersen said...

I should have been more clear: I think we mean different things in 5 pages. 5 pages in LCNS format is ridiculous. 5 pages in HotNets 2-column dense format is actually a lot of information. I'm not sure we're disagreeing about our estimate of how much information belongs in a Hot* paper when you correct for the formatting.

Panos Ipeirotis said...

Please do not forget the workload of the reviewers. We ask people to review 6-12 papers within a period of few weeks, trying to find time between teaching, research, service, traveling, and so on. It would be cruel to allow 50-page long submissions, just because the authors felt that including every single detail and experiment was important.

Having an upper limit in the number of pages forces people to focus on the aspects of the paper that matter most.

I will concede that there are papers that flow better given enough space. But for every paper that improves with extra pages, there are N others (N>>1) that will just cram enough material to reach the upper page limit. (Because if a paper does not reach the upper limit, it is not important enough to be accepted, as we all know.)

Not sure what to suggest, though. I get a feeling that there is something more fundamental that is broken in the academic peer reviewing process, as David says.